Escaping Stability-Plasticity Dilemma in Online Continual Learning for Motion Forecasting via Synergetic Memory Rehearsal [0.03%]
基于协同记忆重演的在线连续学习稳定性和 plasticity dilemma 解决方法在运动预测中的应用研究
Yunlong Lin,Chao Lu,Tongshuai Wu et al.
Yunlong Lin et al.
Deep neural networks (DNNs) have achieved remarkable success in intelligent systems such as autonomous vehicles and robots. However, most DNN-based methods suffer from catastrophic forgetting, where the DNN may fail to maintain its performa...
From Stochastic Conjugate Gradient to Stochastic Second-Order Optimization, Driven by Conjugate Coefficient With Second-Order Information [0.03%]
从随机共轭梯度到带二阶信息的随机二阶优化方法
Zhuang Yang
Zhuang Yang
Conjugate gradient (CG) and second-order information (SOI) receive increasing interest due to their crucial role in improving stochastic first-order (SFO) algorithms for solving machine learning problems. Although a variety of stochastic CG...
Differentiable Clustering Graph Convolutional Network for Hyperspectral Unmixing: Methodology and Benchmark [0.03%]
一种新型可微聚类图卷积光谱解混网络及其基准试验
Mingming Xu,Jin Xu,Zhiru Yang et al.
Mingming Xu et al.
The task of hyperspectral unmixing (HU) is inherently more complex than classification, as it requires separating mixed pixels into pure spectral components, demanding fine-grained spectral and spatial modeling. Traditional convolutional ne...
TAFNet: Trusted Multiview Associative Fusion Neural Networks for Analyzing Dynamic Brain Networks [0.03%]
基于受信任的多视角关联融合神经网络的脑动态网络分析方法
Weiping Ding,Wenhao Dai,Tao Hou et al.
Weiping Ding et al.
Dynamic functional connectivity (DFC) is crucial for analyzing brain networks, as it captures the temporal dynamics of brain regions. However, most existing methods assume uniform quality across time windows, neglecting the inherent data he...
MoADL-TLSTM: Multiobjective Automated Deep Learning-Based Transformer-LSTM for Load Forecasting [0.03%]
基于多目标自动化深度学习的变压器-LSTM负荷预测方法(MOADL-TLSTM)
Kang-Di Lu,Bing-Xu Zhang,Yong Xu et al.
Kang-Di Lu et al.
Accurate load forecasting serves as the critical factor for optimizing energy allocation and ensuring economic operation in cyber-physical power systems (CPPSs). Deep learning (DL) models have emerged as pivotal tools for capturing complex ...
Yupei Zhang,Xian Sheng,Mengfei Liu et al.
Yupei Zhang et al.
Graph transformers (GTs) have recently attracted considerable attention for graph representation learning (GRL). However, the existing methods often neglect hyper-order structures that arise from implicit node-groups within graphs. More cri...
Sliding Integral Neural Network Driven Robust Solution for Time-Varying Quadratic Programming [0.03%]
基于滑动积分神经网络的时间变化二次规划的鲁棒解法研究
Yang Si,Zhibin Li,Kai Zhao et al.
Yang Si et al.
Time-varying quadratic programming (TVQP) requires efficient, accurate, and robust online solvers. Existing discrete-time (DT) recurrent neural networks (RNNs), however, often face a tradeoff between solution precision and noise immunity. T...
Attention-Guided and Role-Aware Reinforcement Learning for Multi-AUV Counter-Game [0.03%]
基于注意力引导和角色感知的多AUV对抗博弈强化学习方法
Wenhao Gan,Kai Guo,Lei Qiao
Wenhao Gan
This article proposes an attention-guided, role-aware multiagent deep reinforcement learning (MADRL) scheme to enhance collaborative decision-making among autonomous underwater vehicles (AUVs) in the counter-game (CG). First, a customized m...
WiCount-DASL: Domain-Adversarial Semisupervised Learning for Wi-Fi-Based Stationary Crowd Counting [0.03%]
基于Wi-Fi的静态人群计数的领域对抗半监督学习方法WiFiCount-DASL
He Wang,Ivan Wang-Hei Ho
He Wang
Wi-Fi sensing provides a privacy-preserving and device-free sensing modality for stationary crowd counting with a low deployment cost. However, labeled channel state information (CSI) data are difficult to obtain at scale, and CSI distribut...
Unleashing the Potential of Imperfect Demonstration for Imitation Learning via Hierarchical Expert Guidance [0.03%]
基于分层专家指导的模仿学习的不完美示范潜能释放
Zhiliang Lin,Zhuangzhuang Chen,Guanming Zhu et al.
Zhiliang Lin et al.
Despite the substantial progress of imitation learning (IL) in training agents to mimic expert behavior, existing methods still suffer from covariate shift and compounding errors due to the limited availability of expert demonstrations and ...